blackyin1 / rbpf_processing

For data collection. detecting objects through change detection and extracting CNN features for tracking in the rbpf_mtt framework

rbpf_processing

For data collection. detecting objects through change detection and extracting CNN features for tracking in the rbpf_mtt framework

Waypoint to visit during the daily routine:

Four times a day, visit the following nodes:

• WayPoint33 - hallway area
• WayPoint34 - master student room at entrance
• WayPoint29 - kitchen
• WayPoint27 - in front of Kitchen door
• WayPoint1 - in front of charging station
• WayPoint9 - meeting room
• WayPoint22 - in front of Jana's office
• WayPoint6 - Nils' room
• WayPoint7 - Rares' room
• WayPoint17 - a bit down the hallway
• WayPoint16 - the end of the hallway
• WayPoint4 - large office at end of hallway

Commands to run to process the data

First, put a symlink in your home folder: ln -s /path/to/data .semanticMap, and then, in order:

rosrun quasimodo_brain metaroom_additional_view_processing -once -notSendPrev -resegment -files /path/to/data -folder /path/to/data

rosrun quasimodo_brain metaroom_additional_view_processing -once -notSendPrev -backwards -resegment -files /path/to/data -folder /path/to/data

rosrun rbpf_processing rbpf_propagation.py /path/to/data

rosrun rbpf_processing rbpf_propagation.py /path/to/data --backwards

rosrun rbpf_processing cluster_objects.py /path/to/data

rosrun rbpf_processing summarize_objects.py /path/to/data

rosrun rbpf_processing extract_features.py /path/to/data

rosrun rbpf_processing reduce_features.py /path/to/data

rosrun rbpf_processing rbpf_conversion.py /path/to/data